The analogy of IT as the fifth utility has just had a frightening reality check. The electrical utility network in the US and Canada that had until now largely been taken for granted simply shut down. Does that mean that our utilities are vulnerable, too? Do we actually want our IT systems to resemble utilities? What about reliability and security? What about your average utility’s byzantine cost structure? Where are the risks? Are there teeth hidden in all those bytes?

Over the next few weeks, we will discuss some of these risks. In this article, we try to nail down just what this phenomenon is, and in follow-up articles we cover what that means, legally. We look at the sticking points in contract negotiations with the vendors, and some of the most important differences in rolling-out such a utility computing system in the US compared with the UK or the European Continent.

The growth of IT since the ‘70s has been a struggle to find standards. That is no less true with this new phenomenon. No one seems knows quite what it is or even what to call it. As Leif Eriksen wrote in his recent article at this site (Will the Real Utility Computing Model Please Stand Up, 2003/07/30), ‘Utility Computing’ means different things to different people. We’ve counted some 14 forms now being pushed into the market. IBM has its “e-business on demand”. HP its “Utility Data Center”. Sun Microsystems its “N1”. Forrester writes about “organic computing” and “fabric-based computing”, and so on.

Trying to bring some order to the confusion of definitions, Mr. Eriksen has boiled it down to “functionality”. He writes that utility computing is the Internet delivery of IT-based functionality on demand, citing among others salesforce.com as an example. The big IT vendors, IBM, HP and Sun et al, don’t put it in terms of functionality. They say they want to deliver storage, networking and processing capacity on demand, whether over the Internet or at your own premises. Both approaches likely involve (at least to some extent) integration, rationalization, virtualization and metering of your local IT facilities, the “customer-specific context”. Presumably you must be in a state to upload as well as to download.

Yet both approaches also involve outsourcing. In his follow-up article carried at this site (Why Utility Computing Will Succeed Where ASPs and Outsourcing Failed, 2003/08/06), Mr. Eriksen distinguished outsourcing from utility computing. He wrote that Utility Computing was “IT-based functionality on demand”, whereas outsourcing was “IT resources, infrastructure, and/or applications for lease”.

But whatever name or price model you may choose, the same fact remains. Some degree of control that you previously had over your IT infrastructure or functionality has been ceded to a third party, either in-house, or outside through a (virtual) private network or the Internet.

As such, the same basic risks are at play under whatever description of Utility Computing you choose to adopt.

There are a number of potential difficulties in the concept of Utility Computing despite the promises of the Utility Computing vendors.

One is the failure for vendors to agree what it is. Is it simply a pricing model, or is it a hardware issue, involving managing and coordinating systems and other resources?

Another is the fact that not all technical elements of Utility Computing are ready for mainstream use. Technical standards that permit disparate internal IT systems to communicate with one another may exist at some of the more basic levels, but are still lacking at the “smart” levels above.

Another danger is the threat of being locked into using a single service provider even though the concept of “Utility Computing” implies that more than one service provider might be the most efficient allocation of service.

We are warned that the outsourcing aspects of the concept may also pose a threat. IT consultant Gartner gave IT outsourcing deals a bad rap in March when it reported that 50% of the deals proved “disappointing”, failing to deliver ‘value for money’. silicon.com just reported (2003/08/21) that the Bank of Ireland faced being "totally closed down" by strikes if the Bank’s staff rejected an agreement negotiated between the unions and the bank regarding a $600m IT outsourcing deal with Hewlett-Packard.

One day earlier, the same newsletter warned of shareholder revolts. It quoted a specialist industry consultant who stated, "Outsourcing is effectively a one-way 'cultural' ticket and the investors believe that they have a right to be heard on the balance this strategically provides against short-term cost savings imperatives."

These are the sorts of risks we will address in coming articles. By knowing where the teeth are and how best to deal with them through contract negotiations, IT executives can avoid falling into some of the same old traps that snared outsourcing deals in the recent past.

Alan Steele-Nicholson is head of the IT-Telecom law department in the Rotterdam, the Netherlands, office of the global law firm Simmons & Simmons, and can be contacted on alan.steelenicholson@simmons-simmons.com. Peter Brudenall is Senior Lawyer in Simmons & Simmons’ London office, and can be contacted on peter.brudenall@simmons-simmons.com. Both are specialists in IT and telecom outsourcing law.