I recently wrote a post on my other blog reacting to a story I had read about a New York consulting firm using Google Apps' Gmail as an emergency replacement mail server for a client whose single in-house server went down with hardware problems. My reaction was surprise at the fact that they apparently went back to the single in-house server once it was repaired, instead of sticking with the far more reliable Gmail or another hosted mail provider.

What I talked about at the end of my post, though, was a technique that I have for minimizing disruptions in making system changes. I'm a fiend for avoiding unnecessary downtime, probably to the point of excess. One of the things that I do from time to time in engagements, depending on the situation, is to plan out a necessary or desirable upgrade and then recommend the client sit on the plans until some (foreseeable) failure rears its head, at which time they take advantage of the situation to not only fix the failure, but also to roll out something newer/better/less failure prone.

It occurred to me as I was writing about it that I've never run across anyone else that does this; I thought I would throw it out there both as a question to see if anyone does, and as a suggestion if they don't. Or am I just crazy?