Chris Hoff has a very interesting post on the hype revolving around recent Cloud failures but reading the post I couldn't find answers for a few but important issues:

The Cloud model of business depends on highly-available infrastructure running software and hardware covered by limited liability warranties. Carbonite may try to sue Promise but anyone that reads the Limited Warranty offered by the vendor, knows that Carbonite will have troubles getting any compensation for their losses, other than hardware costs. It is in fact a funny situation where its SLA with the customer governs service availability while its contract with the technology providers does not offer any sort of warranty.

BTW: The Carbonite lawsuit against Promise is a perfect example on why bugs are negative externalities (and why the "No more free bugs" initiative is indeed a positive shift.)

Another interesting point is the reference to the IBM Cloud Certification Program; certainly one of the weirdest ideas I've ever heard but probably a great marketing opportunity... Let's start with the basic question, who would hire IBM, the Cloud customer or the Cloud provider? And, hey, it is an IT service provider doing auditing others…

If the customer is responsible for the hire, it would be incurring in expenses that would be latter used as marketing by the cloud provider. Not to mention the issue of having an ITO provider auditing its competitors. I can tell you from my own experience that it is possible to audit your competitor on a behalf of a common customer but this is a hell of a weird situation where results are bellow optimal, not to mention the clear conflict of interest…

So, the most probable outcome will be the Cloud provider hiring the certification agent as it currently happens with the ISO standards such as 27001, 9000, etc. Oh gosh, so after the PCI conundrum we are set for the "ISO/IEC XTC for Cloud Computing Quality Management System"? Freaky! VERY freaky! Think PCI! Now think Heartland! And now think RBS! Now try to go sleep…