I came across a fresh report by Information Technology Intelligence that only 15 percent of the 300 enterprise companies surveyed will adopt cloud computing over the coming year. The figures don’t surprise me, nor do I think they should surprise others as well, as plenty of industry surveys have shown similar results.
While Information Technology Intelligence doesn’t provide a sound description of what they’re definition of cloud computing is, judging by the context of the report, it seems they’re going with the typical SaaS = cloud computing equation most have adopted.
Despite its maturity, I still think the SaaS/cloud computing concept is battling some red herring issues that are preventing it from reaching mainstream adoption among the big boys. Security remains one. There’s no reason why cloud-based systems should be less secure than on-premise ones, but the fact remains that they can be. I think many SaaS-only vendors have had some lapses that continue to provide on-premise vendors with an advantage.
In addition, long and high-variance communication latencies, dependency on distributed infrastructure networks, and lack of the hands-on control many CIOs require are still providing resistance.
Finally, I think on-premise vendors are doing a bang-up job with their products in two ways that really matter:
– Low-priced monthly subscription pricing with premium on ease of installation and entry
– Taking a grassroots adoption, one click at a time, based on usability and control as the core advantages.
There are always hidden costs and inconveniences that don’t make it into the marketing hype. In that sense, the current cloud frenzy reminds me a lot of the off-shore software development/service outsourcing frenzy of a few years ago. That “revolution” rapidly shrank to something much more marginalized as the realities of long-distance development and lack of control set in. I imagine that cloud computing and SaaS will continue to experience the same moving forward.
When it comes down to it, SaaS can represent a brilliant application structure in certain niches, but not so much in others. Certain classes of applications and operating environments will never make sense within the limitations of vendor-hosted cloud computing, thanks mostly to the reasons I mentioned above. Yet, still other situations and application sets have such stable usage patterns that it’s hard to imagine how businesses could survive without it.