As more organizations are starting in earnest to consider deploying master data systems, they are also beginning to see where some fundamental issues may block the creation of a full-scale master data repository. This had me thinking about potential ways around some of these issues (especially where governmental regulations prevent moving data across borders!), and one aspect I started considering is the use of a data federation or virtualization model that can abstract the pereption of a unified view without necessarily copying the data.

At the same time, I was approached by Composite Software to do some research on the fesibility of incorporating data quality management techniques within a data virtualization framework. That activity only has continued to pique my interest in integrating virtualization and MDM, and it also allowed me to explore some ideas I explored in my book on data quality management. Meanwhile, the result of that task is an interesting white paper, which you can access via Composite's web site - search down to the "analyst reports" section of the page. Let me know your thoughts!