In a conversation earlier today, someone asked me when I thought there would be no remaining reservations about the cloud. When, I was asked, will everyone stop saying "This has to be in my own data center" or "I only want this behind my own firewall"?

My first response was based on lives and times. "Let's do a thought experiment," I said. "If the typical enterprise IT decision maker is, say, 36 years old, and if anyone who was 12 years old in 1998 has had the Internet as part of their life for as long as they can remember, when will enterprise IT leaders all be people who are totally accustomed to on-line banking? To Web-based email? To Facebook? To a workplace where people use salesforce.com, Concur Expense, and Google Apps?" By simple arithmetic, that day will arrive in 2022.

Then I decided to check myself with something closer to pure math. "I don't know what per cent of the market is still nervous about the cloud today," I said, "but I'd predict that the number will be no more than ²/₃ as large next year, and ²/₃ of that value in the year after that." Like any exponential decay, this never gets all the way to zero, but let's ask how long it takes before 99% of today's objectors are asking "How do I get this done in the cloud?" instead of asserting, "This can't be done in the cloud."

Logarithm of 0.01 to the base ²/₃ is 11.4, which means that 99 out of 100 of today's cloud objectors will be past their tipping point by December 1^st of 2020.

Neither of these is a rigorous projection. I'm not Hari Seldon. But these are two completely independent estimate methods that agree to within about one year of each other, and that makes me feel as if there's some reality here.

Further, working backwards, the same math would suggest that half of today's cloud skeptics will be open to persuasion (if perhaps not actual converts) before Tax Day in April of 2011; that ³/₄ will have reached that state of mind by Christmas Day of 2012.

The trailing quarter of that group will need to have some pretty compelling reasons, it seems to me, to explain why they're still making massive capital investments to enjoy a slower pace of improvement and a longer time to value on their IT initiatives. Why not ask, "How can we make the cloud as good as we need it to be?" An answer that begins with "When, for how much?", will be much more interesting than one that ends with a period after "We can't."