important but not interesting

it's funny how others can be so helpful in summarizing your situation simply by their ability to survey the entire situation from a higher vantage point. i have been fighting tooth and nail for the last month and a half or so with my research, and i could not explain so succinctly what the problem was. not expecting such a clear description of my dilemma, i went and chatted with a well-respected researcher here in the lab a week ago in hopes of gaining some new directions and ideas. the visit was helpful, and i did come away with a positive feel. however, the part that stuck with me the most was his description of my situation, that what i am undertaking is "important but not interesting."

after he said it, i was like 'wow, he perfectly articulated what i have been confusedly feeling for about 8 months now.' the solution to the power crisis is simple: turn sh*t off when it is not in use. the only problem with this is that computing and the internet have been classed as 24/7/365, instantaneous, and never broken. computing and the internet are never supposed to be turned off. the problem with this 'perfect' availability is how it is manufactured using imperfect hardware. this manifests itself as many redundant copies of the same damn data for insurance reliability purposes. not only does all this production hardware have to be on all the time. the back-ups have to be on all the time too. so whatever power is being consumed by the infrastructure serving requests, just multiply that by how many backups you have. crazy right!

i digress......the fact that my solution is to turn stuff off is VERY uninteresting, but VERY important. so......now i am in a situation where i am racking my brain trying to find an interesting spin for a given application space of turning things off and on without anybody noticing but the person responsible for paying the electricity bill.

p.s. - i just couldn't NOT post this link. insane, right. somebody tell the ny ball clubs that we are in a recession