Enterprise Information Management - Reference Book

Exciting new book: Making Enterprise Information Management (EIM) Work for Business: A Guide to Understanding Information as an Asset, by John Ladley, Morgan Kaufmann (3 July 2010), ISBN: 978-0123756954 Read More...

New research questions Symbols and Reasoning

Very much of the work being done today is still on the levels of Symbols and Logic. Take the Semantic Web and Ontologies, for instance. However, there is new research coming out, which questions this. The new direction is more heuristic than logical / analytical. Read More...

The End of the Line for Data Modelling?

Although Data Modelling as we know it has proven useful in many cases (the easy ones), there are certainly limits to the approach. Unfortunately, not being aware of the limits has led to many a disaster - such failed projects, heavy budget overruns, deadlines missed by a year or two, and - in most cases - much too complex logic. Read More...

Semantics vs. Natural Language Processing

I have argued that formal semantics are useful, but they are also, ehm, formal. Which disqualifies them from broad usage in business modelling. But things are getting better. Read More...

New opportunities from atomic data

Working on the most atomic level is a big difference. With “atomic” I mean attributes and attribute relationships. Think Concept Maps, think RDF graphs, think ORM, those kind of things. With “big difference” I mean new kinds of DBMS’s, yes, it is true! And new ways of integration of data, even world wide! Read More...

Trip report from ESWC 2010

Just returned from the Extended Semantic Web Conference 2010, which took place in Hersonissos, Crete, last week. It was a very good event. I wanted to find out, what the actual state of affairs are in semantics. And the answer is: Very good, thank You! There is no doubt in my mind that the semantic web standards are finding many interesting applications. Read More...

Schisma #3 - Logic and Language are different ballgames

If you flip the hat completely and pretend to be a cognitive scientist you quickly realize that the it industry has been applying the right tools to the wrong category of problems. The logic systems (UML, OWL and what have you) work fine in an automation context, but they are incomprehensible in a business environment. Read More...

Schisma #2 - Business Rules!

One of the things you learn when you model information is that sometimes you have to stop drawing, because you get into a lot of details. It could be for instance certain facts about a specific group of products. And other groups of other products have their own specialties as well... Been there? What you do is that you move such details into a list of "Additional rules". So you end up having diagrams and (long) lists of (business) rules. Skill and experience is what makes you stop in time.... In reality the whole diagram is in itself also a representation of a set of business rules!

Fortunately good people from the business rule community (steadfast and sincere people) have been working under the auspices of the OMG (Object Management Group) on something called: Semantics of Business Vocabulary & Business Rules (SBVR). It seems to be much needed. Fortunately it has been developed in such a manner that it integrates with RDF and OWL. Unfortunately we are still missing some good, simple, business oriented tools for visual diagramming of "Business Vocabularies" (what is wrong with Concepts?).

So, maybe the business rules perspective is what it takes to get business concept modelling done right?

By the way SBVR, is also heavily inspired by the ORM modelling technique, which I like a lot. But, which also is too complex for business level analysis and modelling.

Modelling Schisma #1 - Less is more!

There is a mismatch between the needs of a good business information model and the traditions employed by it data modellers. This is a great divide. Read More...