Saturday, August 02, 2008
Information vs data
"Marriage is like a one-way street ... by the time people start telling you to turn around, it's too late to back out." ---- Abbott and Costello
I've been pondering this bit of wisdom. My own hippie-era marriage was a good example. If I had known more about the proper rules for choosing mates beforehand, I would have seen the warning sign and backed out earlier. The experimental data, so to speak, emerged after the damage was done.
And if I had known earlier that termites were in fact common around Spokane, I would have checked the many available sources of data, and would have determined that my "ant problem" was in fact a termite problem. When I bought this house in 1991, friends told me that termites are non-existent here, and even the house inspector told me the same thing. When he found a chewed-up piece of wood under the house, he assured me that it was probably carpenter ants, nothing to worry about, because termites couldn't survive here.
More broadly, the whole process of buying a house and getting a loan is so thoroughly fenced in by discrimination laws and litigation on all sides that realtors and inspectors are effectively required to give only a limited range of data. They are not allowed to give common-sense advice on the characteristics of the neighborhood, because that may require making racial judgments. They are not allowed to ask whether you should be buying a house at all. I suspect this lawsuit-bound reticence is part of the subprime mortgage mess.
We have so much data available today that we've forgotten the importance of information. They're not the same.
Rumsfeld's discussion of known knowns, known unknowns, unknown unknowns, and unknown knowns is widely parodied but valid, and it needs to be expressed in a clearer form, like this: Sometimes you know the proper question and you've already answered it. Sometimes you know the proper question and you haven't managed to find the answer yet. Sometimes you don't understand the situation well enough to form a proper question. ¶ 2:20 PM
I've been pondering this bit of wisdom. My own hippie-era marriage was a good example. If I had known more about the proper rules for choosing mates beforehand, I would have seen the warning sign and backed out earlier. The experimental data, so to speak, emerged after the damage was done.
And if I had known earlier that termites were in fact common around Spokane, I would have checked the many available sources of data, and would have determined that my "ant problem" was in fact a termite problem. When I bought this house in 1991, friends told me that termites are non-existent here, and even the house inspector told me the same thing. When he found a chewed-up piece of wood under the house, he assured me that it was probably carpenter ants, nothing to worry about, because termites couldn't survive here.
More broadly, the whole process of buying a house and getting a loan is so thoroughly fenced in by discrimination laws and litigation on all sides that realtors and inspectors are effectively required to give only a limited range of data. They are not allowed to give common-sense advice on the characteristics of the neighborhood, because that may require making racial judgments. They are not allowed to ask whether you should be buying a house at all. I suspect this lawsuit-bound reticence is part of the subprime mortgage mess.
We have so much data available today that we've forgotten the importance of information. They're not the same.
Rumsfeld's discussion of known knowns, known unknowns, unknown unknowns, and unknown knowns is widely parodied but valid, and it needs to be expressed in a clearer form, like this: Sometimes you know the proper question and you've already answered it. Sometimes you know the proper question and you haven't managed to find the answer yet. Sometimes you don't understand the situation well enough to form a proper question. ¶ 2:20 PM
