Few models within IT have excited as much interest in such a short space of time than Web services. A key ingredient in this interest is due to the widespread support that Web services has received from all the major vendors; with bitter corporate enmities set aside to allow the introduction of open non-proprietary standards. A question that is most often asked by end-user organisations looking to implement this model is: “Will these open standards fragment in the future, and if we make the wrong choice of supplier now will this leave us exposed?”
In the main, and from a technical point of view, I believe that fragmentation will not occur, and the choice of supplier will be based on a different set of metrics than used in the past. This is not say that there will not be some divergence as to the implementation of the standards, but these will be at the periphery, and will not have a deep impact on the chosen technology.
This is analogous to the accepted ‘standard’ of HTML. Widely used and implemented, HTML would be considered by most users as a strong standard, yet there are still differences in implementation different Web browsers. These differences are not so great as to make HTML unusable across different systems, and the same will be the case with Web service standards such as WSDL, SOAP, etc.
Therefore we can see that a major promise of Web services is the removal of choice between expensive systems that enforce proprietary elements and certain restrictions (to a greater or lesser degree) on the implementation of solutions. The Web service model allows for an infrastructure that should remove any restrictions on the implementation of solutions, this leads in turn to solutions that can match the strategic needs of a business without limitations imposed by technology shortfalls.
Reaching this nirvana of having technology working as a totally integrated element of the overall business will not be without its pain points. As vendor organisations look to gain competitive advantage, there will be some areas of Web service ‘standards’ that could possibly be more fragmented. Security is an obvious area of concern within a Web service model, and we are already beginning to see ‘security solutions’ becoming available from different vendors. This is one area where care has to be taken to ensure that any implemented security model is not so tightly tied to a Web service infrastructure that it cannot be modified if the need arises.
Apart from security, the main area of contention surrounding Web services will be the change that it imposes on the way that processes are handled within an organisation. This is so fundamental as to lead to the possibility that some organisations will have to restructure the way that they carry out their business function, even to the point of redefining that function itself.
This derives from the ultimate promise of Web services. Within any organisation there are activities within the implemented processes which are carried out effectively and, conversely, activities that are not so effective. Within any given process it is difficult to distinguish between the ‘good’ and the ‘bad’ activities. Another way to consider this is to find those activities that bring real value to the business and those that cost the business money without the added value.
All the activities will be necessary for the process to function, but with Web services there comes the ability to source activities from a third party. This is most commonly referred to as businesses concentrating on their core processes, but it is in fact more granular than this. Organisations will need to fragment their processes to understand which activities are cost-effective and those that are not. Cost models will then need to be built to show ROI for outsourcing these activities, always bearing in mind that the risk element of relinquishing control of a process part is an important element of the equation.
Similarly, and this is where real business change can come about, those activities that an organisation does well can be externalised as a Web service, and as a source of revenue. The extent to which this will come about is problematic in anything other than a long-term view, but it will happen.
Therefore, the promise of Web services can be summarised as existing in two distinct areas: the removal of technical infrastructure requirements that enforce limitations and the user, and the better understanding of the business as a whole. Of the two, it is the latter that will prove the real force within Web services.
Mike Thompson is Principal Analyst with Butler Group. He can be contacted on 01482 586 149, or [email protected]