When two people communicate they bring to the conversation a wealth of subtle messages and contextual implications. Such interactions are always dynamic -- never the same from one conversation to another. However, when two pieces of software communicate, we strive for fixed, static interfaces in order to facilitate the interaction while abstracting the underlying code -- what we call loose coupling. But in building such contracted APIs, we extinguish any hint of the conversational context so important in human communication.
Such is the challenge for all semantic integration: loosely coupled interfaces, be they Web Services or RESTful APIs, fix the interaction context, eliminating the inherent dynamism of all true semantic interactions. No semantic standard, no traditional integration tool will ever overcome this problem. It's no wonder true semantic integration has always been just out of reach.
The solution to this conundrum is to rethink loosely coupled integration altogether. Add an intelligent agent to the mix -- one smart enough to resolve differences in interaction context between disparate software endpoints dynamically and in real time. Far more than a traditional broker, which must rely on static transformation logic to resolve endpoint differences, this semantically-aware agent must be able to interpret semantic metadata, as well as policies, rules, and the underlying data themselves to create real time interactions that maintain the semantic context -- what we call dynamic coupling, a central principle to Agent-Oriented Architecture.
Attendees of this session will learn:
- The limitations of loose coupling and why we need dynamic coupling to address them
- Why the battle between loosely coupled interfaces and semantic integration will never end without a rethink of integration
- How an Agent-Oriented Architecture approach can resolve the semantic challenges inherent in loosely coupled interfaces like Web Services and RESTful APIs.