I was recently invited to a G-cloud “Quick Wins” session put together by RedPepper52 which, sadly, I couldn’t attend. The cloud is something that has been on its way for some time and is starting to gain traction.

(As an interesting aside, check out this article from 2005 in which Nicholas Carr predicts the future. Truly visionary)

I expect that Nicholas Carr’s vision will come true in the next decade and we will wonder why we ever had internal IT departments. This is interesting for two reasons: firstly, the move itself requires a different kind of technical architecture and IT provision mentality, and secondly IT managers need to become more focused on exploiting the computing resources on offer rather than provisioning them.

But a private cloud for Government? Is this really a likely scenario?

The agenda for the quick wins seminar suggested that commodity services which require little configuration would be the first to go over to this new setup. I’m not sure I agree. Most IT departments in local government have email systems and they work just fine, the capital is already invested and they are run at reasonable cost. Moving these to the cloud would incur transition costs, there may be some issues with local application integration (our apps send emails) and (perhaps most importantly) there is no added value. So we pay a load of money for a small cost reduction in future (if that even materialises) and the increased risk of losing local expertise.

No way.

To my mind, the cloud quick wins lie in providing new stuff. Stuff we haven’t got yet. XRMs. Master Data integration platforms. SOA middleware.

Actually, SOA middleware is the killer app for the cloud. Few councils make extensive use of it and it provides a way to pool services and construct meaningful applications without having to invest in a lot of in-house expertise. We can build our G-cloud architecture one service at a time and use these back-end web services to provide apps on a multitude of existing and emerging end user devices.

If the G-cloud is our future, we can’t do “quick wins”. We have to build from the back-end.