Event Processing

TIBCO has launched a new product called TIBCO BusinessEvents
Which they say "proactively identifies and identifies complex correlated events" wonder how that works.Currently it is available on Windoze and Solaris platforms
Guess they will soon have to bring out something that works on Linux as well !
They have great products but it is difficult to find some good technical literature on how the products work atleast at a high level.
In that aspect i consider the DB Server Concepts Manual from Oracle is a great leader, and the fact that they have made the documentation open on technet is just wonderful.

Alan Alda's interviewing tips

I got directed to this Intelligent Enterprise article from Mark Rittmans blog-post.
It sure makes for some interesting reading, The quick study section summarized by Ralph Kimball, Margy Ross says it all for anyone gathering requirements for a DWH/BI project.

Amazon selects Cognos

Amazon.com has selected Cognos -Report Net for their enterprise-wide reporting requirements, More interesting is the fact that it will be a Linux based setup.
Cognos news

Turning data into action for Customer Insight

In an article on sas com John Gaffney and Larry Dabrow defines customer insight as "Customer insight is where customer value and customer needs intersect. It is where data is transformed into understanding."
They further go on to mention that to get true understanding and retain existing customers enterprises need to get data from two places "The first source is the data from transactions, demographics and attitudes. The second source, which many companies ignore, is data about customer needs and values. It is this fusion, however, that results in the insights that lead to smarter decisions and higher profit."
Another key observation they make : "all available customer data must be deployed fully, because you can bet that the competition will deploy it."
Too often i find that enterprises in their data warehousing integration efforts "Reject" lot of records because of some data quality issues , Gleaning information for action from these "Reject" cases is a challenging task for any data-modeller , Lot of times it is just very easy to error out these records and proceed ahead with the routine , These error cases are supposedly sent back to the business for validation but in my experience business is just not bothered...wonder if we can get some statisitcs of opportunities lost because of this...