Quality Evaluation Models - High Level Models and Methodologies for Information Systems - page 95

multiplicity of lines, illustrating the different levels of performance of all the eight
goals.
Each of the eight proposed goals has some relationship to each of the four
aspects of system

environment interaction that we have previously described. In
that sense, extendibility and security are the goals of a successful boundary defi-

-

-
nition, as the purpose is allowing for useful entry into the system as well as
protection from harmful entry. Flexibility and reliability are features that permit the
system

s internal structure to adapt and survive in the environment. Functionality
and usability will allow for effectors of the system to successfully perform, and
connectivity and privacy are essential aspects for the receptors to manage the
analysis of the environment so as to use it

'

in the system ' s maintenance and

development (Whitworth et al. 2006 ).
The area of the web translates the performance of the system on the whole, so a
larger area will mean a potentially more robust system. The shape of the web is a
portrait of the system

' s performance. This shape varies according to the speci c

features of the system s environment, which might require higher levels of per-

formance from some goals. The shape of the web will highlight those variations.
(Whitworth et al. 2006 ).
Ideally, this web should be balanced. While different systems may determine
different shapes, in a general sense, it can be af

'

rmed that a very steep performance
increase in one direction of the model may not imply the system

s success if it also
brings about a very steep decrease in another dimension. So, for example, if a
system has very high flexibility but very low reliability, it will most likely not
perform well (Whitworth et al. 2006 ).
It is important to note that in the WOSP model, the concept of performance,
where the system

'

is success and quality is determined, is not an absolute concept, but
rather it is viewed exclusively in relation to the system

'

' is environment (Whitworth

et al. 2006 ). This implies that there is no single de nition to determine system

performance, because every system will be different when put into context. This
fundamentally alters the paradigm set out with the TAM. The TAM was an extre-
mely simplistic model because it aimed at being useful regardless of the system

' s

context, whereas the WOSP model is entire focus is precisely in the possible vari-

ations that can occur when context (or environment) is taken into account.
This aspect of the system is demonstrated in Shore and Zhou

'

s( 2009 ) evaluation
of the virtual environment Second Life using the WOSP approach. The authors
de

'

ned each of the goals contemplated by the WOSP model in a manner that would
suit the peculiarities of Second Life. All the criteria were used as the model
requires, but each of them was assessed and adapted to the speci c context of

Second Life. This method allowed the authors to reach interesting conclusions
regarding numerous aspects of the system, such as Second Life flexibility and

reliability due to common server lag or downtime, but very high connectivity and
extendibility, thanks to an extensive system of user-created content (Shore and
Zhou 2009 ).

' s low