David Stodder has a great expert analysis at Intelligent Enterprise, which considers the two recent announcements in the MDM world (Informatica's purchase of Siperian and Talend's announcement of its open source MDM product). I left this comment over there:

"One aspect to consider is the fact that the desire for implementing master data management is usually a symptom of deeper data management issues that have not been resolved. The promise of a "golden copy" or "single source of truth" belies the need for higher quality entity data. The problem is indeed as suggested in David's article: the people buying the solution have not thought out the scope of the work involved in transitioning an *enterprise* (with its scores, if not hundred or thousands of independent applications and spreadbasemarts) to committing to the use of a single "master data asset." So you end up with siloed master data implementations (which is an oxymoron if I ever heard one) that serve smaller workgroups. (Psst: really, that MDM implementation is YAID - yet another inconsistent database).

 Talend does have an interesting strategy, particularly from the perspective of the clean slate. If your organization is flexible enough to assemble an enterprise strategy and then build towards that strategy, the "developer" approach actually might be good for maintaining a consistent, unified view of the critical business concepts through which the applications execute the business.

 So I have started to see some of our clients think about MDM from a different point of view. Instead of thinking about it as the single data asset for the business, they are looking at it as a tool to achieve the objectives that inspired the need in the first place - consistency, accuracy, high quality customer records. In turn, they are using the master data services to help filter out bad data in the first place, which essentially "bakes" the governance into the applications (the same way error handling is built into software)."

Meanwhile, I dealt with a lot of the questions that David Stodder has raised in the book that I wrote on Master Data Management, which apparently is being sold at a 48% discount at Amazon.

I was scanning the news this morning, and came across a few interesting notes regarding impending acquisitions of master data management vendors:

IBM to buy Initiate Systems? Also, some sources mentioned an all-hands meeting scheduled for next week.

Informatica to buy Siperian? Also, a source commented to me that a senior staff member at Siperian who was there for 7 years has moved on to a new role at another company.

This Forbes interview of Chevron CIO Denise Coyne suggests that a 2010 focus for the oil and gas giant is data quality, although the code words employed scream Master Data Management:

"... we're going to create a pilot enterprise project to consolidate all of that information in one place."

"We have lots of data about people in one organization, another database about people in another organization. Consolidating that information to have one source of the truth, to be able to make faster, more competitive decisions more quickly, is a really important focus in 2010."

Meanwhile, my book on Master Data Management is now (Jan 20) on sale at 51% off the cover price, and I hope someone at Chevron buys one for Denise Coyne!

In a move that we might have predicted last year (with the increasing chumminess between the two companies from Q2 and on of 2009), Oracle has purchased data quality company Silver Creek Systrems. Great move for Oracle - SIlver Creek Systems has solid technology that has been focused on the product data quality niche, and the acquisition adds another component to Oracle's already firmly-established offerings for master data management.

Apparentyly, the same issues that plagued competing US intelligence agencies immediately after the 9/11 bombings have not yet been resolved. According to this Time Magazine article, President Obama's summarized the failure to prevent terrorism suspect Umar Farouk Abdulmutallab from boarding a Detroit-bound plane was that "The U.S. government had sufficient information to have uncovered this plot and potentially disrupt the Christmas Day attack, but our intelligence community failed to connect those dots."

Yet again, we see that despite being flooded with data, there was a failure to turn that data into actionable knowledge. Apparently, according to the article, inteligence agencies knew that the suspected bomber Abdulmutallab had traveled to Yemen, a spot of brewing anti-US terrorism plots, that his father had contacted the US embassy in Nigeria to notify them of his son's activities, that no one asked whether Abdulmutallab had a US visa, or whether he should have been added to the no-fly list. Also, the fact that he purchased a one-way ticket and no checked luggage might have raised some concern as well.

Any of these events should have triggered some action, but the fact that they didn't potentially raises a different question: how often do we miss events that should trigger a security response? I am sure a lot more frequently than we'd like to believe, and that might raise your level of anxiety.

And that raises another different question: what is the probability/risk that a missed event is a critical one like the Dec 25th situation? Of course, a low probability might alleviate some of the anxiety.

However, from a data perspective, the issue is a matter of data sharing and integration - protocols for capturing the key semantic aspects of logged events could be published to a common repository that could be continuously monitored, mined and evaluated to determine when some proactive action should take place. Is MDM the answer? Maybe, or perhaps a master repository published to a cloud environment with layered data services for rapid identity resolution...

Oh, check out this interview to understand a little more about national security.

I hope everyone had a great Thanksgiving weekend, and are ready to get back to work for the home stretch of the year. I had an interesting experience last Wednesday - my daughter came hoem from school and mentioned that her science teacher had emailed the class letting them know that presuming that the weather was clear, one could view the international space station followed by the undocked space shuttle atlantis cross the sky.

At 6:11 PM EST on Wednesday, I went outside with the kids and we stood out there with our heads looking almost straight up into the (somewhat cloudy) sky. The little ones kept pointing at airplanes, but all of a sudden we saw two bright objects appear, one following the other, sail across the WNW sky, then duck behind some clouds and disappear.

As a kid, I used to excitedly watch the launchings of the Apollo mission Saturn rockets, wondering whether the astronauts would reach their destination or even return alive. But with the multitude of space junk orbiting the earth, and with limited exposure to how incredible it really is to get big things off the ground and floating around the planet, today's children are somewhat immune to the wonder of space flight. Yet it is amazing that my daughter's science teacher can reintroduce that wonder by giving us the chance to experience it first hand, in our own front yeard.

Not really a surprise, and one that we discussed over at this blog note, but apparently the data reported regarding stimulus money spending is not immune to data flaws. This article over at cnn.com discusses some of the simple types of errors appearing in the jobs data, such as hundreds of thousands of dollars spent in Arizona's 52nd, 15th and 86th congressional districts, despite that sparse state's only having eight congressional districts.

I especially like dthe quote from Wisconsin representative Dave Obey: "Credibility counts in government, and stupid mistakes like this undermine it."

I am presenting a talk today in a web seminar at noon ET. The talk is on "Best Practices Using Core Data Services," and I will explore how we can organize our critical information management and business intelligence activities on top of key data principles for quality, availability, and enhancement!