I have worked with dozens of I.T. groups, and their behavior is startlingly consistent: they're not about to risk changing anything. And it's for a simple reas

I have worked with dozens of I.T. groups, and their behavior is startlingly consistent: they're not about to risk changing anything. And it's for a simple reason: they are measured on Not F*cking Things Up. They are not about to have everything fall apart on their watch, because they honestly believe they will not have a job. Realistically, I think the only way business I.T. will ever change, is for the metrics to change. Rather than the stupid and threatening measure of "don't screw up, ever, or else" and these 99.999999999% uptime numbers, I.T. must be guided under a new principle: "evolve constantly, at a reasonable pace, taking acceptable risks". What's frustrating to me is that they "freeze", wanting to delay change as long as possible, not appreciating the irony that a snowball is being created. If they wait 5 years to upgrade a browser instead of 1 or 2, the cost of a mistake is far higher. If they refuse to allow beta software, ever, they are losing lead time on detecting possible issues. The list goes on.