This falls into the category of what might be referred to as “purpose-built PaaS” – IBM is introducing a platform based on an existing set of software services that to this point had been offered only in a packaged model. We love having IBM-sized marketing dollars pumped into educating a large portion of the market in the capabilities of SaaS / PaaS, but we’re not in love with the concept of PaaS as an extension of a core service offering…a la Salesforce.com’s Force.com PaaS offering. On the one hand, integration is built in and users will already be familiar with the functional aspects of the system, on the other hand if every major vendor starts offering PaaS to simply extend what their packaged software does, we’re going to have dozens of islands of PaaS that will hurt the concept and the market as a whole. This article is spot-on in its assessment that Big Blue does hold an advantage in selling to an existing customer base, but at the end of the day, vendor-specific PaaS defeats the core value proposition of PaaS from the perspective of creating additional management overhead, forcing organizations to compromise on business process design because of integration and platform capability limitations, and requiring technical resources to learn yet another PaaS / SaaS skillset. The goal with PaaS is to achieve (relative) ubiquity, not more of the same old and tired IT thinking repackaged as a hosted service.
