I have mentioned before the natural, or unnatural perhaps, proclivity of Technology writers to go a step too far. Utility Computing is another good instance of this phenomenon. I can't think of another group, or market, or industry that always takes the argument and goes flying by where everyone else is. Zoli made a laugh out loud comment about it by declaring Web 4.0. So I am not the only one amused by the polemics. Utility Computing is now the topic of much conversation. Nicholas Carr loves this idea. For a person who makes his living trying to figure out the IT future it probably makes sense to pursue this idea. If nothing else he raises some very interesting questions about why corporations have so much unused computing capacity? On the other hand, Vinnie Mirchandani has seen Utility Computing up front and personal and he has some legitimate questions about what's possible and why we have not started to see some of the benefits of utility computing from the IT giants (who are as big as utilities). Both of these writers make some good points, but I have to say that neither seems on the mark really. I guess that perhaps I have a more pedestrian imagination, but when I think of utility computing I think of NetSuite and other software as a service vendors who offer companies an experience very similar to the on-premise client-server model - custom configuration of a system to match your business model - but hosted on a multi-tenant architecture that enables huge economies of scale when it comes to hardware and personnel resources. Carr, especially, seems to envision a world where the transactions of thousands are handled on a single system, metered like electricity or water. Perhaps there will be some of that, but I will bet that we will not do away with the idiosyncrasies of business models - and the custom configurations that they require - any time soon. One of the things that neither writer brings up, and they are both at least as old as I am so that can't be an excuse, is that though the corporate computing resources, both the servers and the 1000s of PCs, are under utilized they are still much less expensive than what they replaced, real human beings doing white collar drudgery. The first time I used a personal computer, mid 80s, the instructor was a women who spent most of her life at General Motors as a financial analyst. She was literally legally blinded by the experience. She had the thickest pair of glasses that I've ever seen. She described her life in GM finance as years of green ledgers and adding machines putting together sales forecasts. The desktop spreadsheet programs made her and thousands of others unnecessary, for GM, that is. She was thrilled with this turn of events and had made herself a spreadsheet expert. So even though there are $2000 computers that are utilized at under 20% they are still incredibly less expensive than the live financial analyst with an adding machine and an unlimited supply of green ledgers. Carr is correct that companies will eventually move towards a utility computing model but it is not the one he envisions, I believe. Corporations will continue to require a custom configuration of their business model in the software. NetSuite and other software as a service vendors are in the right place in the market for this level of utility. | 
