So I do a ton of work for government clients (in fact, looking back I can't think of a single non-government agency I've worked for since arriving) and Juneau being the state capital that means I also do some work for the federal government.

I just finished up a contract for a client that involved moving tons of data around and combining it and finding trends (in an awful Database environment to top it all off).  Basically this information is used by this agency to 1) advise legislators and 2) help the agency make decisions that could very easily impact people's livlyhoods, as in, put them out of business.

I'm a data person, give me a set of data and I can mash it, tear it apart, and put it all back together to mean something.  I can also look at a set of data and say 'you've put it together way wrong'.  This client's information though was not only bad ('we can't get an accurate count of this so we estimate and fudge') but it is assembled in ways that are impossible to reproduce even by the guy who did it the first time.  The data doesn't even represent the same thing now as it did 5 years ago in some cases, still, its mashed and graphed and the agency is patted on the back.  One day someone is going to wise up and say 'give me your data and your algorithm so I can verify you destroying my business' and they will not be able to do it.

I knew governments were inefficient, but this just kinda boggles my mind.  It makes me wonder what the heck people are thinking when they say 'more government oversite on this' or 'why doesn't the government...' I mean, c'mon, really?  This is basic college level data analysis (maybe high school, its been a while) and the government doesn't get it a little wrong, they got it so wrong I'm struggling with a way to describe it.

PS.  I love my job, I really do, you have no idea.