The cold war over who owns the data in enterprise may be about to run hot. There is always tension business executives and the minions of the chief information officer over who has ownership of the data that makes the company run.

Business executives feel strongly that they should have ownership of the data that reflects their business operations. Of course, the CIO at the behest of the CFO and CEO is always trying to get to the one truth about the overall business that should be reflected in the data. All too often, however, business executive roll out "new" spreadsheet numbers that better reflect an argument they are trying to make than the numbers that have been collectively rolled up by the central systems managed by IT.

Usually, the data being presented by a business executive is derived from a spreadsheet running on a local PC or a database running on a departmental server. But in recent months a number of vendors, including LogiXML, Kalido, LucidEra and Good Data have been rolling out business intelligence applications as a services positioned as tools that can be easily adopted by end users.

A big part of the pitch surrounding these services focuses on the fact that they give business executives access to more powerful tools to analyze their business without requiring the dedicated intervention of IT. On one level you can easily see what these might be attractive to the business executive and even IT people that don't want to have to go to the trouble of supporting a business intelligence application.

But Chris Rafter, a vice president of consulting services at Logicalis points out that many of these business intelligence as a service strategies fly in the face of the spirit of good data governance. Instead of encouraging all data stake holders to commit to creating one truth about the state of the business, these applications are encouraging business executives to create even more complex data fiefdoms.

Unless customers have developed a robust middleware strategy for rolling up data that resides in these applications, Rafter argues that companies would be better served by bucking down on data discipline to the point that no data that resides outside of the corporate business intelligence system is allowed to be relevant during any given meeting.

That may require more discipline than most companies are capable of today. But when you look at the compliance regulations that are starting to evolve around data governance, it may quickly become apparent that companies will have no choice but to agree to a single business intelligence application standard.

Clearly, there is a lot of frustration with business intelligence applications today. In environments where there are no business intelligence applications, services can make a lot sense. But before people run off to solve that problem by bringing in another system, it might behoove us all to consider whether the cure is ultimately going to be worse than the disease.