eWEEK content and product recommendations are editorially independent. We may make money when you click on links to our partners. Learn More.
A mid-July google search on the phrase “Web services” returned almost 2 million hits. A follow-up search, however, on the singular “Web service” returned one-fourth that number. In short, the idea enjoys enormous buzz, but concrete description of an actual example seems to be harder to find—even though the ideas involved are actually far from new.
The promises now called Web services were being made more than a decade ago in forums such as the 1988 Object-Oriented Programming Systems and Languages conference in San Diego. Visionaries there spoke of an object marketplace in which application feature lists would be determined by users rather than by vendors. A generic framework, with basic functions such as document composition, would incorporate additional functions—developed internally or purchased from object builders—to create a tailored enterprise application.
Objects would combine data structures with the code required to manipulate them, avoiding the brittle coupling of traditional applications.
Substitute XML for open data structure, SOAP (Simple Object Access Protocol) for uniform messaging protocol and ubiquitous HTTP over TCP/IP in place of the bandwidth miracle assumed in those 1988 keynote speeches, and youre where we are today: able to build and deploy these modular applications but still with no idea of how a remote object market will work—and with little idea of how to test applications that may never run the same code twice and whose modules we do not own.
We can avoid disappointment by not overdefining what “Web services” means. In the enterprise, we can realize many of those late-1980s visions with Web services formalization of interfaces and freedom from proprietary protocols. With suitable care in defining access privileges, we can even build federated applications that streamline supply chain interactions.
These are useful things to do, and we dont need any fairy dust to sprinkle over the problems of establishing trust on public networks. Web services are a viable application architecture today, even if they wont be a robust and vigorous marketplace until many days after tomorrow.
What are your organizations Web services plans? Contact me at peter_coffee@ziffdavis.com.
Related Stories
- Concentrate on the Core Technologies
- W3C Releases Web Services Schemes
- Keeping Web Services Royalty-Free
- Web Services Security: A Political Battlefield