Sometimes the disconnect between media hype and reality is almost palpable. For over a year now we've been using the term "cloud computing" to describe a Venn diagram of complementary trends. God knows that 'cloud' is a better tag than SaaS or other descriptors for this unwieldy crew of services - especially if you have to come up with hundreds of punning headlines for a living -- but the overwhelming fact that I have been forced to confront is that not a lot of people know what the 'cloud' is, and many of the minority of people who think they do have some very odd ideas about it. And finally, a fair number of these people are CIOs or others paid to come up answers at some point.

I've probably heard the views of over 1000 people on the subject and it's clear to me that we in the media have sped off way down the road and, having seen the blueprint for a brave new world, we've gone ahead and put up the signs and organised the party, assuming that incoming tenants will sort themselves out. The problem is that there's no guarantee that they will and we could be left with a tattered mess of ill-defined sprawl.

There are plenty of numbers around but given the state of some of the research I'd rather report some ad hoc opinions.

Definitions of cloud computing: There are loads of them but precious little exists in the way of agreement. Some people use 'cloud' to mean anything connected to the internet; others seem to use it as a proxy for 'datacentre'. Broad interpretations bracket services that make no good sense to not do over the internet (WebEx, message security scanning, antivirus) with siloed applications reached over the web (Google Docs or GMail) or platform-as-a-service approaches such as's and fully-fledged (fluffed?) public cloud environments such as's AWS and Microsoft's Windows Azure. And then of course there are...

Private clouds: Cloud computing not complicated enough for you? Try bringing in internal environments that take on some characteristics of public clouds such as metered billing or support for line-of-business chargeback but keep systems in-house for security/governance/other reasons and therefore lose benefits such as economies of scale.

Why you would do the cloud thing: If your business is in crisis - say you run a media business or the UK government - then it's virtualy the only game in town for slashing IT costs and it's tempting to bet the farm. Alternatively, if you need to move fast and just have to go now without untold budget to throw at the problem, the cloud is attractive. It also looks nice if your business is spiky and you don't want to pay for a lot of under-utilised kit. And it's also a quick win and sop to CEOs and CFOs as you save on admins, software licences, servers and datacentres.

Why you wouldn't do the cloud thing: Concerns over who gains access to sensitive data, especially abroad. Worries over stability of suppliers. Flaky network connections. Pesky regulators with rulebooks drafted in tablets of stone relating to where data sits and pesky legal departments that have read ancient charters. You're secretly scared by the possibility of losing the IT department or you're even more scared by possibility of losing my own job. You've been so busy that you haven't got a clue what people are talking about.

This, it seems to me, is where we are with cloud computing -- only just starting out. Perhaps by the time journalists and analysts have started to flag the real interest will begin.