Google's blog goes into a lengthy explanation of the outage and how Google tried to solve it. I have to say, though, that the explanation makes me feel as if cloud computing is a more tenuous proposition than I realized. Consider this explanation of what happened, according to the Google blog:

The Gmail engineering team was alerted to the failures within seconds (we take monitoring very seriously). After establishing that the core problem was insufficient available capacity, the team brought a LOT of additional request routers online (flexible capacity is one of the advantages of Google's architecture), distributed the traffic across the request routers, and the Gmail web interface came back online.

The key point here is that the Gmail team knew about the failue within seconds, but it still took 100 minutes to bring Gmail back online. This isn't to criticize the Gmail team --- such outages, I believe, are inherent in the nature of cloud computing. There are simply too many points at which things can go wrong --- not just servers, but to the Internet itself.

And that's why Microsoft is laughing all the way to the bank. Enterprises simply can't have outages of their most vital applications. All Microsoft needs to do is arm its Office sales force with news reports of the outage. When they're up against the Google sales team trying to sell Google Apps, all the Microsoft people will have to do is pull out the news reports. Chalk up one more sale to Microsoft.

Microsoft will be moving Office to the cloud to a certain extent, but that will most likely be an adjunct to client-based Office. And until cloud computing is a rock-solid certainty -- a time which may never come -- enterprises will most likely opt for reliability over cost-savings.