Also try not to take ownership of wikipedia pages. Whatever you write will evolve in strange ways once other people start editing it. There’s lots of crap that ends up in wikipedia. For the pages that interest me I visit them every year or so and remove the worst of the damage. You could of course poll an article regularly and try and amend it after every change, but that seems like a lot of work.

As for cloud storage, I have not yet investigated the Amazon SimpleDB (the only real cloud storage I know off the top of my head). Lots of other companies have cloud storage offerings that seem like S3 or a CDN – both of which seem to fall short of the full “cloud” idea. Let me know if there are any particular areas you think I should investigate (I lack the time to fully investigate it at the moment).

Thanks for your comment.

I was asked to comment on your post by a journalist on the grounds that it’s critical of the Wikipedia article I created about cloud computing, and specifically excludes storage. I spent quite some time researching the various viewpoints in order to create what I still think is a reasonably good consensus as to what cloud computing is about.

First thing’s first, from a user-centric point of view cloud computing is (as you observe) about abstraction and hiding complexity; users consume computing resources without knowledge or concern about the systems responsible for delivery. Some of these happen to be centralised (eg most SaaS providers who use large grids to deliver their applications) and others are distributed (eg seti@home, skype) – people have tried to equate cloud computing with centralised or distributed computing but the user doesn’t [need to] care whether the service is delivered from a datacenter or other clients. Skype for example is hybrid in that it promotes certain clients to be ‘supernodes’ (servers) and supports this network with centralised billing, authentication, POTS gateways, etc.

Regarding your comment about storage, the same could be said for other components of cloud computing; it is the maturity of the various components that has triggered the cloud computing ‘revolution’. Raw storage is one thing, but combined with compute, web services, queues, etc. one can create some far more interesting architectures – and the two areas you cite as most interesting (infrastructure and platforms) are essentially useless without storage.

Thanks for contributing to the discussion,

Sam