Last week in this space, I questioned U.S. CTO Vivek Kundra's vision for helping the government achieve computing efficiency, I questioned whether the various branches and agencies of the U.S. government could ever cooperate well enough to take full advantage of the promise of the cloud. I highlighted interagency strife and security and confidentiality issues as big obstacles.

Moments ago, influential blogger Malik Om questioned whether providers were yet ready to meet the demand for SaaS. Om wrote, "If, like our little company, you run your business using Google Apps, you’re playing with fire. For time and again, the company has proven that despite all its talk, its offerings are as unreliable as those of any other service provider." You can see his whole blog at http://bit.ly/xD4rs.

Like most people, I am a frequent user of Google's free services, especially search. Om, however, says that his company is a paid user of Google Apps. I'm sure reliability is part of what they expect.

In similar past circumstances, providers have blamed failures on a number of arcane systems. Unexpected router overload, especially during routine maintenance, seems common.

Moving critical applications requires much deliberation. Potential users, and readers of this blog, know that power and cooling systems have to be well designed. Readers of other publications know the strength and weaknesses of Internet connectivity and software needed to keep cloud apps available. Still there are any number of other potential pitfalls to prematurely moving mission critical applications.