| While we cannot with any certainty predict what the future holds for Web services, there do seem to be some commonsense observations that we can offer. The first of these is that the Web services community by and large consists of a plethora of small vendors offering technologies and services at the bottom end, and a handful of large vendors (dominated by IBM and Microsoft) at the top end of the scale. Given the sheer number of smaller vendors (and the prevailing economic conditions), we expect there will be a degree of consolidation with the larger vendors using their financial muscle to add smaller vendors' niche technologies into their portfolios. On a technical level, we expect that a number of technologies, which largely follow the technologies we have seen in this book, will rise to prominence: XML and XML Schema will become the fundamental building blocks of all enterprise technology and will remain the basis of Web services throughout their lifetime. SOAP 1.2 and WSDL 1.2 will prevail as the messaging and protocol-description languages of Web services, respectively, though at the time of this writing, SOAP 1.2 is a new recommendation and WSDL 1.2 remains a work in progress which means that toolkit support for both is currently limited. UDDI will increase in its use for inter-enterprise Web service discovery and integration purposes. The use of UDDI at runtime for system reliability and failover will also increase. An eventual standard for transactions will likely be influenced to a great extent by the WS-Transaction standard. While we think it unlikely that WS-Transaction will remain in its current form, any future transaction standard will share a number of its characteristics. It is unlikely that BTP will become a widespread transaction protocol simply because its list of backers is somewhat less prestigious and powerful in the Web services arena than those of WS-Transaction. WS-Security, as well as its associated WS specifications, will emerge as the dominant standard for securing Web service environments. As IT organizations become more familiar and more experienced with securing Web services, the adoption of Web services by enterprises as well as the release of Web services from within the firewall to the outside world will increase. Conversations, choreography, and orchestration are becoming increasingly intertwined. It is anticipated that a single standard (either based on BPEL4WS or perhaps as the output from the W3C's choreography working group) will subsume all of these and so conversations and workflow will become one. Web services that are accessible by mobile devices will steadily increase. In the near term, most mobile systems will be based on proxy architectures (e.g., using J2EE servlets as the proxy) instead of direct Web service invocations.
Generally, we expect the near-term future to be punctuated by both continued work in the specification space and for software engineers to begin projects that utilize basic aspects of this technology (like SOAP and WSDL). If all goes well, then by the time that we as engineers are ready to tackle more advanced features, the standards groups will have completed their own work and we should find ourselves in a position where the technology is mature and engineers will have already scaled the learning curve to be able to use it. All of this means that Web services have a bright future ahead of them. |
