Loraine Lawson asks a great question: "Can Semantics Tech Eliminate the Need for Data Integration?"
"I'm a lost cause, it seems, so I might as well fess up to the fact semantic technology fascinates me, particularly when you move beyond the over-hyped 'Web semantics' stuff and look at real applications..."
Loraine is not alone, it's a common question. Good time to chime in on this.
Semantics are not integration, they are simply a way we represent information within information systems. In fact you can consider semantics as the meaning of data and the use of data, simple put. I've defined it within the context of integration as a mapping between an object/attribute, represented and stored in an information system.
Thus within the world of data integration when one system is exchanging information with another, you have to deal with the differences in application semantics, this typically means a semantic mapping mechanism. Almost all integration servers have them. However, the concept of semantics goes much further.
When considering data integration you have the semantics of the source system, and the semantics of the target system, but there is also the concept of semantics defined as a common base of understanding between many connected systems. Thus, we not only have the concept of mapping native application semantics from one system to another, but the notion of mapping native application semantics to a common semantic understanding of the integration domain. In my EAI book written oh so many years ago I called this the common data model, or CDM, but the underlying concept is the same.
So, what semantic tech can do is to create a common semantic understanding of an integration domain, but typically does not provide the mechanisms for integration unto itself, including extraction from the source, transformation, translation, routing, transport, and production to the target. Hope this helps.
Good post. Consultants tend to oversell whatever they are selling and lately I've heard a few pitches that seem to assert that the right ontology or semantic technology is the cure for whatever might be the customer's integration pain. There is a lot to be said to having a solid semantic understanding and agreement regarding data, processes, etc., but as your post points out, there's more to integration than a good semantic foundation.
Agree with the points David is making. As it regards expressor, we don't see our semantic metadata abstraction layer as a cure for all DI problems, but we do believe that creating a common data layer is very important for a variety of reasons, including the ability to create reusable business & transformation rules and to improve data governance in a variety of ways.
Michael Waclawiczek
VP, Marketing
expressor software
www.expressor-software.com
I totally agree !
Extending the semantics abstraction to cloud computing, would then require a 'Common Data Model', that are globalized (and agreed upon). This in turn would require a common semantic ontology (see RDF.org) agreed on every subject matter Domain on top of a common Domain ontology (almost similar to Esperanto and Volapyk, these artificial languages which never gained success, by the way).
Is this the way we are heading with OASIS, Dublin Core etc.?
Interesting to hear opinions ..
Part of the problem is that there are any number of collections that could all carry the label 'semantics' (application semantics, infrastructure semantics, business semantics, database semantics, etc) and they are often so bound up into one another that nobody bothers to try to untangle them.
I would bet a dollar to a dixie cup that most enterprises care more about the everyday global semantics of business than all the others combined.
I'm working on a framework (called Q6) that provides businesses with a predictable, extensible and completely persistent abstraction layer that can make all of their information (data and content) interoperable with any other collection. Sounds ambitious, but it is being used in a couple of soon to be released applications.
Nicknamed 'Quantum Semantics' it, like chess (and all languages), is based a small number of constraints that when consistently applied can manage an infinite number of combination, permutations and scenarios.
Stay tuned!
"Quantum Semantics" - sounds intriguing. What would you call a "quant"? I am developing a similar idea also based on small number of constraints and resulting logical space (www.ontospace.net).
Care to elaborate?