So I'm sitting here at Google's Atmosphere conference in London among a well-heeled business crowd that the search giant wants to attract to Apps and other wares, and I'm wondering whether I'm a believer in cloud computing or not.

In a professional sense, I want to be. Being a journalist you always hope that the disruptive technology will succeed because it allows you to write a man bites dog story. Any crack in the Microsoft machine is a headline and the internet computing is certainly the biggest threat so far to the client/server gravy train.

In an empirical sense, I'm halfway convinced. You can't run away with fingers in ears from the success of a, for example, but looking outside of areas such as CRM/salesforce automation, employee performance management, email, collaboration and some elements of security, business cloud computing is still a niche, even a niche of a niche. Think of the big bastions: ERP, word processing, spreadsheets, presentations... the programs still sit on disks. Vendors hate being confronted with this but security and governance remain huge concerns for many prospects and in many cases the products are not full-featured or are flaky.

In terms of the future, I'm kind of convinced. Web-based competition is growing sharply, albeit from a small base. Listen to people like Rentokil, Valeo, Telegraph Media Group or universities and colleges. They are prejudiced in favour of the cloud because they tried it and it worked for them. It fits industries in crisis (the media, for example), the cash-strapped (education, startups), those who need to move fast (for testbeads and new projects) and those who like opex over capex.  

The truth is that this is a complicated argument that has been reduced to a cartoon. The very term 'cloud computing' is used as a cover-all (for applications, storage, processing, conferencing, even subscrition tariff software) so you get useless polls that tell you x per cent of people use it but y will use it inside three years. Private clouds morph with public clouds and everybody has some data on who's doing what. And they don't mean a damned thing.

Cloud computing is set up as a revolutionary alternative to client/server when for most large companies it will be a complement for decades to come. The numbers regarding cost are played fast and loose and the whole thing becomes a bunfight.

We need to take the emotion out of this and accept that cloud-style approaches will become pervasive but don't necessarily mean the death of the IT department, the end of Microsoft (already a large provider in the space) or the collapse of civilisation.