Gens gave a lucid answer: "I'm not worried about the blurring line between hosting and cloud. Hosters will use more cloudlike architectures to get more efficient, and cloud guys will add more SLAs, options for private lines, etc., becoming more responsive to traditional enterprise needs." At that point, Gens backed into IDC's definition of cloud computing: "If hosters add utility (shared) services, consumption-based pricing, self-service, and APIs -- four big things most don't yet do -- then they'll be 'clouds.'"
Gens says these four criteria also define the private cloud -- and that in surveys IT management at large enterprises prefer the private cloud to the public cloud for security, availability, performance, and cost reasons. No question, says Gens, that public cloud offerings will ultimately have the most disruptive effect. But private clouds will be a necessary interim step in large companies unwilling to cede control of runtime infrastructure to providers.
Personally, I wonder how many big IT departments will be able to justify the level of effort to make their "services" meet the four-part criteria Gens articulated. If you ask me, growth of public or private clouds may appear to be dramatic next year, but that's mainly because current levels of adoption are so low.
But that's just my reaction. The IDC Web site lays out 10 predictions for 2010 that you can agree or disagree with.