Information Technology Reference
In-Depth Information
A SaaS-ES-solution is overall less flexible than on-premises ES in that the ES-
adopter can not completely customize or rewrite its code. Because of this, the
SaaS-ES-adopters must be prepared rather to change their business process to fit
the solution than to align it to enterprise requirements. Also such adopters often
face massive coordination effort because they have to integrate hosted software
from various vendors with their existing ES solutions and/or legacy applica-
tions. How to run an effective RE process for SaaS-ES projects is by and large
unknown. We, however, think that further research efforts in this direction are
warranted, because SaaS-ES solutions represent an important development in the
field.
5 Conclusion
This chapter surveyed the requirements elicitation and modeling approaches in the
sub-area of RE for ES. We reasoned about some tacit assumptions these approaches
make and why these assumptions might not be realistic in all ES contexts. Based on
this we derived directions for future research. We acknowledge that such a survey
can bring only a snapshot view on a fast-changing area. However, we think some
lessons can be derived from it.
First, RE-for-ES has a long future ahead. ES will stay, though the on-premise
ES solutions will have to live with new types of ES, namely FOS-ES and
SaaS. The context of these projects gets increasingly more cross-organizational
on both the ES adopters' side and the ES vendors' side. That the ES adopters
are cross-organizational businesses calls for developing cost-effective approaches
for handling requirements for business coordination. ES solutions include hosted
and on-premise ES modules provided by multiple vendors, and this calls for
cost-effective approaches to the complex problem of aligning the coordination
mechanisms embedded in multiple packages to the coordination requirements of
the ES-adopters. The elicitation and modeling approaches developed in the RE
community in the past decade might only partly serve the needs of the ES projects
embracing the current market trends.
Second, our analysis gives us enough evidence that ES implementations have
impacted RE research regarding sub-areas as requirements elicitation and model-
ing. This means that RE researchers (active in non-ES project contexts) who design
solutions to problems in those sub-areas should evaluate their proposed solutions
regarding how they work in the ES context. In general, if a solution proposal is
meant to be industry-relevant, then researchers have to evaluate and generalize its
usefulness in various contexts. We think that ES is one significant context, for which
such validation evaluations should take place.
Third, we witness that the majority of RE-for-ES techniques have been devel-
oped and evaluated by means of empirical research methods. This alone is an
achievement, given the inherent difficulties in carrying out this type of research
activity.
Search WWH ::
Custom Search