Home Experience Contact Other Info Discussion
Everything as a Service

I have been intrigued by the various trends that have swept through the IT industry over the years. Just about every single one has promised to reduce the cost of ownership, increase the speed of delivery, and allow businesses to concentrate on their core purpose. We have seen outsourcing, insourcing, and offshoring. Open source has barged onto the scene, and proprietary systems have fought back. Processing power has been pushed out to the desktop, and now virtualisation is re-inventing the mainframe. Client devices have been thick and thin, data has been personal and corporate, and databases have been centralised and distributed. No wonder businesses are confused – the IT industry doesn’t seem to have a consistent view on how to deliver IT services!

The latest model seems to be “Everything as a Service” – and that’s a term that Hewlett Packard is already using. “Software as a Service” (SaaS) is widely available, at least in some environments, and “Platforms as a Service” (PaaS) are emerging. Expect to hear about Hardware, Computing, Security and Middleware as a Service, if you have not already come across them. Could these be the magic bullets that sort out every business’s technology problems once and for all?

The answer could be both “yes” and “no”.

Why “yes”? Research consultancies are beginning to acknowledge that there are real benefits in using these models, as long as it is done in the right way, and for the right reasons. These technologies are fully capable of delivering what they promise, and they can be a very effective way of supplying a service. But they don’t run the business, and they can’t make fundamental business decisions – and that’s where the “no” applies.

It may be stating the obvious, but IT should only exist in a business to help the business to operate more effectively or more efficiently. It is also true that some businesses are so dependent on computer systems that they couldn’t function in the absence of technology, but even in such organisations, the technology in isolation is not the business. It is an enabler. Hopefully, the business will exist long after today’s (and tomorrow’s) technology has moved on.

What is important is the separation between the WHAT and the HOW. Once you know WHAT you want your systems to do, you can consider the options for HOW they will do it – and then you should expect this to trigger further thoughts about the WHAT. Deciding what to buy will almost certainly be an iterative, circular process, rather than a linear one, and throughout the process you need to remain focused on the costs, the benefits, the risks, and the impact of whatever change you are contemplating. At some point you will be ready to make a decision... and when that time comes, don’t delay. Make that decision, and then get on with implementing whatever you have chosen. Just be sure that you know what you are committing to, before you sign on the bottom line. And don’t follow fashion just because it is fashionable: make an informed decision, be flexible, and make your technology work for you.

Iain Millar, 27 January 2011