re: Why IT Has A Credibility Problem
You only make these mistakes if you delude yourself to begin with. Take "the cloud". No one in their right mind thinks it's a good idea to hand all your data to Microsoft instead of keeping it on your own server. For no reason. And PAY them for it.
The same with virtualization. CPUs are cheap, virtualization is complex, fragile, and costs money with no ROI. And the same with software-as-service, VOIP (using your computer as a telephone), mobile apps (using your telephone as a computer), and all the other "new paradigms that will revolutionize IT!"
It's like in software engineering. Beginning with structured programming, every few years some new way of doing it was supposed to "design the bugs out!", be "self-documenting!", and "multiply programmer productivity!" I fell for it until I wised up.
We don't get fooled again.
Just keep looking at how the users do their jobs and how they could do them faster and easier with [fill in the blank]. Adding a spell-checker to the secretary's word processor passes that test. Making her work with the "ribbon" interface she hates on a virtual desktop through a touch screen using Windows 8/Metro running on a computer in Redmond instead of one under her desk does not.
Every product that every vendor comes up with is supposed to "revolutionize IT!" The only things that actually did were relational databases in the 60's, PARC's mouse-driven GUI in the 70's, the PC in the 80s, and networks in the 90's (including the internet). Other than that, it's mostly a lot of marginally-improved and expensive hot air, hoo-hah, and ballocks.
-- faye kane