Last month I interviewed the UK Government’s Chief Information Officer John Suffolk for a software-as-a-service supplement that appeared as an insert to the Guardian on July 2nd – unfortunately it is not online at the moment. The question was how the public sector might take advantage of cloud computing.

A few days later, the Times revealed that the conservative party is contemplating a policy which includes storing personal health records with Google or Microsoft. This idea, and the links between the Tories and Google are further examined by John Lettice here.

The Tories are not in government. Nevertheless, I was fascinated by the contradiction between what Suffolk told me, and this new proposal. Suffolk:

What matters to us is first of all, where are the data centres located? What’s the scalability and the security? Do I believe in using somebody else’s cloud, where I don’t know what they’re doing in it? No, I never foresee that if we’re using personal data.

He is talking up the idea of a government cloud built on the public sector network – in other words, an independent implementation.

Although this idea sounds reassuring to those who dislike the idea of handing over our personal data to Google or Microsoft (and I am one such), it lacks the immediate benefit and cost-saving of replacing internal systems with an existing cloud provider.

Why sink all that public money into reinventing what is already available, and possibly ending up with something that works less well?

The themes are familiar. How secure is the cloud? How reliable? Can we trust Google? How much are we willing to pay, for the greater freedom and control that comes from owning our own systems?

It is not implausible that a year from now the UK may have both a Tory government and a desperate need to curb public spending. In the end this is going to be an economic as well as an ideological argument; and I suspect it will be increasingly difficult to be high-minded about putting personal data on “somebody else’s cloud”.